Sets Math 130 Linear Algebra
Total Page:16
File Type:pdf, Size:1020Kb
get away with \and so forth." If you want to see all of what \and so forth" entails, you can read Dedekind's 1888 paper Was sind und was sollen Sets die Zahlen? and my comments on it. In that arti- Math 130 Linear Algebra cle he starts off developing set theory and ends up with the natural numbers. D Joyce, Fall 2013 The real numbers. These include all positive Just a little bit about sets. We'll use the lan- numbers, negative numbers, and 0. Besides the nat- guage of sets throughout the course, but we're not ural numbers, their negations and 0 are included,p fractions like 22 , algebraic numbers like 5, and using much of set theory. Still, it would be useful 7 to know a little bit about it. transcendental numbers like π, e. If a number can A set itself is just supposed to be something that be named decimally with infinitely many digits, has elements. It doesn't have to have any struc- then it's a real number. We'll use R to denote the ture but just have elements. The elements can be set of all real numbers. Like N, R has lots of oper- anything, but usually they'll be things of the same ations and functions associated with it, but treated kind. as a set, all it has is its elements, the real numbers. If you've only got one set, however, there's no Note that N is a subset of R since every natural need to even mention sets. It's when several sets number is a real number. are under consideration that the language of sets becomes useful. Subsets. If you have a set and a language to talk There are ways to construct new sets, too, and about elements in that set, then you can form sub- these constructions are important. The most im- sets of that set by properties of elements in that portant of these is a way to collect some of the language. elements in a set to form another set, a subset of For instance, we have arithmetic on R, so solu- the first. tions to equations are subsets of R. The solutions to the equation x3 = x are 0, 1, and −1. We can Examples. Let's start with sets of numbers. describe its solution set using the notation There are ways of constructing these sets, but let's S = fx 2 R j x3 = xg not deal with that now. Let's assume that we al- ready have these sets. which is read as \S is the set of x in R such that The natural numbers. These are the counting x3 = x." We could also describe that set by listing numbers, that is, whole positive numbers. 1 is the its elements, S = f0; 1; −1g. When you name a first natural number, 2 the second, 3, the third, etc. set by listing its elements, the order that you name We'll use N to denote the set of all natural num- them doesn't matter. We could have also written bers. Some people like to include 0 in the natural S = {−1; 0; 1g for the same set. numbers, but I follow Dedekind who started with Open and closed intervals in R are also subsets 1. There is a structure on N, namely there are op- of R. For example, erations of addition, subtraction, etc., but as a set, it's just the numbers. You'll often see N defined as (3; 5) = fx 2 R j 3 < x < 5g [3; 5] = fx 2 R j 3 ≤ x ≤ 5g N = f1; 2; 3;:::g which is read as N is the set whose elements are 1, 2, 3, and so forth. That's just an informal way of de- There are a couple of notations for subsets. We'll scribing what N is. A complete description couldn't use the notation A ⊆ S to say that A is a subset 1 of S. We allow S ⊆ S, that is, we consider a set DeMorgan's laws describe a duality between in- S to be a subset of itself. If a subset A doesn't tersection and union. They can be written as include all the elements of S, then A is called a proper subset of S. The only subset of S that's not A \ B = A [ B a proper subset is S itself. We'll use the notation A [ B = A \ B A ⊂ S to indicate that A is a proper subset of S. (Warning. There's an alternate notational con- The distributivity laws say that intersection and vention for subsets. In that notation A ⊂ S means union each distribute over the other A is any subset of S, while A ( S means A is a proper subset of S. I prefer the the notation we're (A [ B) \ C = (A \ C) [ (B \ C) using because it's analogous to the notations ≤ for (A \ B) [ C = (A [ C) \ (B [ C) less than or equal, and < for less than.) Products of sets. So far we've looked at creat- Operations on subsets. Frequently you deal ing sets within set. There are some operations on with several subsets of a set, and there are oper- sets that create bigger sets, the most important be- ations of intersection, union, and difference that ing creating products of sets. These depend on the describe new subsets in terms of previously known concept of ordered pairs of elements. The notation subsets. for ordered pair (a; b) of two elements extends the The intersection A \ B of two subsets A and B usual notation we use for coordinates in the xy- of a given set S is the subset of S that includes all plane. The important property of ordered pairs is the elements that are in both A and B: that two ordered pairs are equal if and only if they have the same first and second coordinates: A \ B = fx 2 S j x 2 A and x 2 Bg: (a; b) = (c; d) iff a = c and b = d: The union A [ B of two subsets A and B of a given set S is the subset of S that includes all the The product of two sets S and T consists of all elements that are in A or in B or in both: the ordered pairs where the first element comes from S and the second element comes from T : A [ B = fx 2 S j x 2 A or x 2 Bg: S × T = f(a; b) j a 2 S and b 2 T g: As usual in mathematics, the word \or" means an Thus, the usual xy-plane is R × R, usually de- inclusive or and implicitly includes \or both." 2 The difference A − B of two subsets A and B of noted R . a given set S is the subset of S that includes all the Besides binary products S × T , you can analo- elements that are in A but not in B: gously define ternary products S × T × U in terms of triples (a; b; c) where a 2 S, b 2 T , and c 2 U, A − B = fx 2 S j x 2 A and x2 = Bg and higher products, too. There's also the complement of a subset A of Sets of subsets; power sets. Another way to a set S. The complement is just S − A, all the create bigger sets is to form sets of subsets. If you elements of S that aren't in A. When the set S is collect all the subsets of a given set S into a set, understood, the complement of A often is denoted then the set of all those subsets is called the power more simply as either A or Ac. set of S, denotes P(S) or sometimes 2S. These operations satisfy lots of identities. I'll just For example, let S be a set with 3 elements, S = name a couple of important ones. fa; b; cg. Then S has eight subsets. There are three 2 singleton subsets, that is, subsets having exactly one element, namely fag, fbg, and fcg. There are three subsets having exactly two elements, namely fa; bg, fa; cg, and fb; cg. There's one subset having all three elements, namely S itself. And there's one subset that has no elements. You could denote it fg, but it's always denoted ; and called the empty set or null set. Thus, the power set of S has eight elements P(S) = f;; fag; fbg; fcg; fa; bg; fa; cg; fb; cg;Sg: Functions and function sets. A function f : S ! T from a set S to a set T can be identified with its graph. Its graph is a particular subset of the product set S × T , namely, the subset f(x; y) j x 2 S and y = f(x)g: You can tell which subsets A of S × T are graphs of functions. They're the ones with the following property: for each x 2 S, there is exactly ordered pair in A whose first element is x. It's convenient to identify functions with such graphs. All the functions from S to T can be collected together to form a set, sometimes called a function set, and denoted either T S or F(S; T ). Since each function is a subset of S × T , this function set is actually a subset of the power set of S × T , that is, F(S; T ) ⊆ P(S × T ).